Member of Technical Staff, Performance Optimization

What you'll do

• Profile and optimize GPU, kernel, and system performance, including latency, throughput, memory use, and compute efficiency.
• Implement low-level optimizations and scale training and inference across multi-GPU, multi-node environments.
• Build performance benchmarking and monitoring infrastructure and collaborate on hardware-efficient model execution.

What they ask for

• Bachelor’s degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent practical experience.
• Proficiency in CUDA or ROCm and experience with GPU profiling tools.
• Familiarity with PyTorch and performance-critical model execution.
• Experience optimizing performance or high-performance computing systems, distributed-system debugging in multi-GPU environments, and understanding GPU architecture, parallel programming, and compute kernels.

Nice to have

• Master’s or PhD in a related field; experience optimizing large models for training and inference.
• Knowledge of compiler stacks or ML compilers, open-source ML/HPC contributions, cloud-scale AI infrastructure, or hardware-aware model design.

Pay

$175K – $220K USD base salary

Tools

CUDA, ROCm, Nsight, nvprof, CUPTI, PyTorch, Triton, XLA

AI focus

Optimize AI infrastructure for LLM, VLM, and video-model training and inference, including GPU kernels, quantization, mixed precision, model graphs, and hardware-efficient execution.